 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
76.127.54.176
In Reply to: RE: And yet... posted by fmak on March 09, 2021 at 03:51:31
to convert USB to I2S, as did the earlier DENAFRIPS Terminator. The newer Terminator (which I also have) uses a firmware conversion method.
I have not compared USB to I2S directly using the Terminator as it's in a differing system but if I hear the same results and the I2S direct is better than AES/EBU via balanced cable on the DENAFRIPS, I might sell the dCS Network Bridge and go all Pi.
Follow Ups:
There are Amanero boards and Amanero boards. The issue in question is whether they use the convent garden XOs or decent ones for conversion into playable resolutions.
Here it is. Best picture I can find on the net.
As it does not look like the one that was listed on the AMANERO site a few years back when the DAC was new, I'm guessing Audio-GD is/was using a licensed version or licensed it themselves and had it made.
Look at the XOs. They are non descript, the brands may vary between the x44 and x48 XOs and even the usb one. Perhaps 20cents each?
Fred, when did you become a component engineer and USB interface designer?How's your massively noisy Dell "Power PC" music server performing?
Edits: 03/10/21
Unlike you I don't post nonsensical comments
Actually you do, you just don't realize it.
Two chips on the ANAMERO board.
Larger one by Atmel (a Silicon Valley company) and it appears to be a SMART ARM-based Flash MCU. DigiKey wants $7.42 in each quantities.
Smaller chip is by Xilinx, another Silicon Valley California company, and it appears to be a XC2C64A CoolRunner-II CPLD (complex programmable logic device) something like an FPGA. DigiKey gets about $3.80 for single quantities.
I doubt if the price of either has much to do with the quality of the AMANERO board as with most things at this level of digital its all being done in firmware and these two chips are all about programing.
You can't tell which ones are the XO chips? They are not the ones you posted about. They are the ones nearer the top by the capacitors. They control frequency accuracy, stability, and phase noise - central to good SQ.
No wonder you have not been able to comprehend the simple truths about XOs used in sample rate conversions over so long..
The DENAFRIPS Terminator in my main system, where I normally run the dCS Network Bridge for streaming, uses CRYSTEK CCHD-957 type clocks and USB > I2S conversion using FPGA. If i's all about cheap clock, it should sound every bit as good with using USB input from the Pi4 as I2S over HDMI (which the Terminator also supports).
But still, while the difference is small, I2S over HDMI still wins. And the Pi3 sounds better using AES3 via BNC.
"No wonder you have not been able to comprehend the simple truths about XOs used in sample rate conversions over so long.."
You seem to like to use acronyms rather than spell things out, Fred, which is why I missed your earlier point.
Since you prefer acronyms, see if you can figure this one out.
GFY!
Quite simple. The method of clock transmission and regeneration at the dac are different. You said earlier that spdif sounded better than usb. At the spdif interface, there is s phase lock loop that aims to reduce jitter. The width of the pll affects SQ as well as other things. The usb interface may or may not have a Femto clock but its design and the interface with the dac also affect SQ.
On top of this is that most China designed units do not implement a Sound Control Panel where such things as latency can be set to fine tune SQ. In my experience, high buffer settings actually 'kill' the transient character of the music. The Topping units now seem to buck the trend.
In my experience, high buffer settings actually 'kill' the transient character of the music.
Agreed. I got a worthwhile SQ boost by setting the buffer size on my Xmos-based WaveIO USB> I2S board to the lowest possible value.
++++
**Really obscure tweak warning**
If your music server's and/or NAA's NIC allow it, try lowering their buffer sizes too.
You heard it here first . . .
D
I have actually been paying attention to how buffer sizes affect SQ for quite a few years, going back to CMP times.
I wonder why no one appears to have flagged and discussed the subject before.
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
